metadata
configs:
- config_name: default
data_files:
- split: training
path: data/training-*
dataset_info:
features:
- name: image
sequence:
sequence:
sequence:
sequence: float32
- name: label
sequence:
sequence:
sequence: float32
splits:
- name: training
num_bytes: 896538640
num_examples: 5
download_size: 44501267
dataset_size: 896538640